<h1 id="qwen-img2img-ultra-grand-final">🔥 QWEN IMG2IMG ULTRA GRAND FINAL</h1><hr /><h2 id="overview">✨ Overview</h2>This workflow is a high-quality Img2Img pipeline built around Qwen Image, designed to push:<ul><li>🎯 image fidelity</li><li>🧠 prompt understanding</li><li>🧩 structural consistency</li><li>🎨 final visual quality</li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8ded0085-2de6-4a69-ae53-b9ec70d7373b/width=525/8ded0085-2de6-4a69-ae53-b9ec70d7373b.jpeg" />🎯 Workflow goals✔ maximum image control ✔ excellent prompt interpretation ✔ premium visual output ✔ powerful and clean upscaling ✔ modular &amp; easy customization<hr /><h2 id="use-cases">🎨 Use cases</h2>This workflow is perfectly suited for:<ul><li>🧍 realistic images</li><li>🎭 stylized art</li><li>🧑‍🎨 character refinement</li><li>✨ image enhancement</li><li>🎬 cinematic renders</li><li>🔍 controlled img2img generations</li></ul><hr /><h2 id="what-this-workflow-does">⚙️ What this workflow does</h2>This pipeline includes:✅ Qwen Image as the core generative model ✅ Full Img2Img workflow with strong structure preservation ✅ Smart denoise control for the right balance between creativity and fidelity ✅ Ultra-sharp final upscale using SeedVR2 ✅ Optional face detection &amp; segmentation for cleaner results ✅ Clean node logic — easy to read and modify ✅ Optimized for stability and repeatability👉 This workflow is designed to produce consistent results, not random outputs.<hr /><h2 id="why-this-workflow-stands-out">🏆 Why this workflow stands out</h2>✔ exceptional prompt comprehension ✔ better composition consistency than classic SD pipelines ✔ strong identity preservation ✔ cleaner textures ✔ improved line stability ✔ fewer upscale artifacts ✔ more predictable results🎯 Ideal for creators who want quality + control, not chaos.<hr /><h2 id="required-models-(direct-downloads)">📦 Required Models (Direct Downloads)</h2><hr /><h3 id="qwen-image">🧠 Qwen Image</h3>UNet GGUF ➡️ Recommended: Q8_0 (best quality / speed balance) 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/unsloth/Qwen-Image-2512-GGUF/resolve/main/qwen-image-2512-Q8_0.gguf?download=true">https://huggingface.co/unsloth/Qwen-Image-2512-GGUF/resolve/main/qwen-image-2512-Q8_0.gguf?download=true</a>🔹 Lightweight alternative (lower VRAM): 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/unsloth/Qwen-Image-2512-GGUF/resolve/main/qwen-image-2512-Q4_K_M.gguf?download=true">https://huggingface.co/unsloth/Qwen-Image-2512-GGUF/resolve/main/qwen-image-2512-Q4_K_M.gguf?download=true</a>VAE 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/Qwen/Qwen-Image-2512/resolve/main/vae/qwen_image_vae.safetensors?download=true">https://huggingface.co/Qwen/Qwen-Image-2512/resolve/main/vae/qwen_image_vae.safetensors?download=true</a>(Fallback if needed) 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/Qwen/Qwen-Image/resolve/main/vae/qwen_image_vae.safetensors?download=true">https://huggingface.co/Qwen/Qwen-Image/resolve/main/vae/qwen_image_vae.safetensors?download=true</a>Text Encoder / CLIP 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors?download=true">https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors?download=true</a>🔹 FP16 version available in the same repository if required.<hr /><h3 id="seedvr2-upscaler">🚀 SeedVR2 Upscaler</h3>DiT GGUF (sharp version recommended) 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/cmeka/SeedVR2-GGUF/resolve/main/seedvr2_ema_7b_sharp-Q4_K_M.gguf?download=true">https://huggingface.co/cmeka/SeedVR2-GGUF/resolve/main/seedvr2_ema_7b_sharp-Q4_K_M.gguf?download=true</a>🔹 Other quantizations available here: 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/AInVFX/SeedVR2_comfyUI/tree/main/gguf">https://huggingface.co/AInVFX/SeedVR2_comfyUI/tree/main/gguf</a>VAE 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/AInVFX/SeedVR2_comfyUI/resolve/main/vae/ema_vae_fp16.safetensors?download=true">https://huggingface.co/AInVFX/SeedVR2_comfyUI/resolve/main/vae/ema_vae_fp16.safetensors?download=true</a><hr /><h3 id="face-and-mask-models">👤 Face &amp; Mask Models</h3>YOLOv8 Face Detector 👉 <a target="_blank" rel="ugc" href="https://huggingface.co/Bingsu/adetailer/resolve/main/face_yolov8m.pt?download=true">https://huggingface.co/Bingsu/adetailer/resolve/main/face_yolov8m.pt?download=true</a>SAM — Segment Anything Model 👉 <a target="_blank" rel="ugc" href="https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth">https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth</a><hr /><h2 id="models-installation-path">📁 Models Installation Path</h2><pre><code>models/
├─ unet/
│ └─ Qwen/
│ └─ qwen-image-2512-Q8_0.gguf
├─ vae/
│ ├─ qwen_image_vae.safetensors
│ └─ ema_vae_fp16.safetensors
├─ clip/
│ └─ qwen_2.5_vl_7b_fp8_scaled.safetensors
├─ gguf/
│ └─ seedvr2_ema_7b_sharp-Q4_K_M.gguf
├─ sam/
│ └─ sam_vit_b_01ec64.pth
└─ ultralytics/
 └─ face_yolov8m.pt
</code></pre><hr />💪CREATED AND BATTLE TESTED WITH 🚀This workflow runs rock-solid and ultra-fast on my beast setup:- CPU: Intel Core i9-14900KF- GPU: ASUS RTX 5090 ASTRAL- RAM: 128GB DDR5 8000MHz Corsair Vengeance- Motherboard: ASUS ROG Maximus Z790 Extreme- PSU: ASUS ROG STRIX 1275W Platinum- Case: Hyte Y70 Touch- Cooling: Asus Ryujin III Extreme 360 ARGB (Noctua edition) AIO Watercooling- Storage: 30TB NVMe M.2 (high-speed array)Zero crashes, buttery smooth generations &amp; upscales even at high res / batch sizes.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/38df00b0-aceb-44a6-b3f0-e6bb66e34886/width=525/38df00b0-aceb-44a6-b3f0-e6bb66e34886.jpeg" /><hr />😎 Casual Tested With My 2nd Config 😎Yeah, it runs smooth on my older beast, no need for a 5090 to enjoy this workflow!- CPU: Intel Core i9-9900K- GPU: ASUS TUF NVIDIA RTX 3080 Ti- RAM: 128GB DDR4 5200MHz- Motherboard: ASUS TUF Z390-PLUS GAMING (WiFi)- Cooling: NZXT Kraken X63- PSU: Gigabyte 750W 80+ Gold- Storage: 8TB NVMe M.2Expect lower speed generations and executions.<hr /><h2 id="important-notes">⚠️ Important Notes</h2><ul><li>🔧 Designed for ComfyUI</li><li>💾 Recommended GPU: 12–16 GB VRAM or more</li><li>🖼️ Final results strongly depend on input image quality</li><li>🔄 Always restart ComfyUI after installing new models</li><li>🆓 Workflow provided as-is, free to use and modify</li></ul><hr />⚡Personal BackEnd Setup For the 1st and 2nd Config 🚀😏My Backend Setup- Python 3.10.6- PyTorch 2.7.0 + CUDA 12.8- Flash Attention 2 (custom build)- SageAttention 2.1.1- xFormers (custom build)- DeepSpeed 0.16.5- Triton Windows 3.3.0.post19- torchao + torchsde- huggingface_hub with hf_transfer (for super-fast downloads)<hr /><h2 id="message-from-the-creator">❤️ Message from the Creator</h2>I spent many hours building, testing, refining, and optimizing this workflow with one clear goal:👉 Make it powerful, stable, and accessible to everyone.If this workflow helps you, inspires you, or saves you time, I would truly appreciate it if you could:<ul><li>❤️ leave a like / buzz</li><li>⭐ follow my Civitai page</li><li>🎨 share your creations made with this workflow</li></ul>Seeing what you create with it is honestly the best reward.👉 <a target="_blank" rel="ugc" href="https://civarchive.com/user/EKKIVOK">https://civarchive.com/user/EKKIVOK</a>Thank you for your support 🙏 — EKKIVOK<hr />⚠️ Important Disclaimer &amp; Responsible UseThis workflow is provided for creative, artistic, and personal experimentation purposes only.- Do not create or generate images of real people without their explicit consent.- Do not use it for deepfakes, non-consensual explicit content, impersonation, harassment, deception, or any harmful/illegal purpose.- Respect copyright, trademarks, privacy laws, and all applicable regulations in your country.- When sharing generated images publicly (including on Civitai), clearly indicate they are AI-generated and follow Civitai's [Content Rules &amp; TOS](<a target="_blank" rel="ugc" href="https://civarchive.com/content/tos">https://civarchive.com/content/tos</a>).By using this workflow, you agree to use it in a lawful, ethical, and responsible manner.Thank you for keeping AI creation positive and respectful ✨

Files with SHA256: eff670564cfdd69c3c673ce62f6c4d95cc41d387d689ba7c4b3a1952896a27da